[Cryptography] [FORGED] Re: How programming language design can help us write secure crypto code

[Ray Dillinger]

On 10/29/2015 08:58 PM, Peter Gutmann wrote:

> __attribute__(( nonnull 1 )) \
> int double( int *ptr )
>   {
>   if( ptr == NULL )
>     return( -1 );
>   return( *ptr * 2 );
>   }
> 
> [...]
>   thing = double( thingPtr );
> [...]

I have never used that annotation in earnest yet, but
I'll give it a guess....

You claimed to be passing it a pointer to an integer,
with an annotation that told it that a null pointer
was an error.

Therefore gcc developers will claim it is entitled to
assume the pointer is Non-Null. (or rather, that the
program has no semantic requirements in the case where
the pointer is non-null).

Therefore I'm guessing it elides the test for null and
its consequent return statement as dead code, then crashes
(if you in fact HAVE called it with a null pointer)
when it tries to dereference the pointer in order to
multiply it by 2.

Am I wrong?

			Bear



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: OpenPGP digital signature
URL: <http://www.metzdowd.com/pipermail/cryptography/attachments/20151029/2fb2f6de/attachment.sig>